Tianqing Zhu

This paper systematically revisits and expands the threat model for backdoor attacks on semantic segmentation, proposing a unified framework (BADSEG) that demonstrates severe, previously overlooked vulnerabilities in current and emerging segmentation models.

The paper introduces ARES, a novel and practical gradient inversion attack that reconstructs sensitive training samples from large batch updates in Federated Learning without requiring architectural modifications.

The paper proposes Two-stage Backdoor Hijacking (TSBH) to create persistent, trigger-activated malicious behaviors by manipulating the observable Chain-of-Thought (CoT) process in Large Language Models.

The paper proposes Cluster Segregation Concealment (CSC), a novel defense that identifies and neutralizes backdoor triggers by relabeling poisoned samples to a virtual class, achieving near-zero attack success rates with minimal accuracy loss.

WebTrap introduces a stealthy, mid-task hijacking attack that successfully compromises browser agents during long-horizon tasks by seamlessly fusing malicious instructions with the original user goal.

The paper introduces CAESAR, a novel multi-agent framework that coordinates LLM agents across five specialized roles to improve success rates and stability in complex, multi-stage cyber intrusion tasks.

The paper proposes Safety Context Injection (SCI), an inference-time framework that prepends a structured external risk report to protect Large Reasoning Models (LRMs) against sophisticated jailbreaks, significantly reducing attack success rates.

This paper investigates safety dynamics in the use of Role-play AI Companions through interviews and a 14-day assessment, identifying key factors shaping these dynamics and revealing short-term emotional relief masking longer-term deterioration.